From: Florian Weimer <fweimer@redhat.com>
To: libc-alpha@sourceware.org
Subject: [PATCH] elf: ld.so --help calls _dl_init_paths without a main map [BZ #27577]
Date: Mon, 15 Mar 2021 11:03:28 +0100 [thread overview]
Message-ID: <87ft0wybrj.fsf@oldenburg.str.redhat.com> (raw)
In this case, use the link map of the dynamic loader itself as
a replacement. This is more than just a hack: if we ever support
DT_RUNPATH/DT_RPATH for the dynamic loader, reporting it for
ld.so --help (without further command line arguments) would be the
right thing to do.
Fixes commit 332421312576bd7095e70589154af99b124dd2d1 ("elf: Always
set l in _dl_init_paths (bug 23462)").
Tested on i686-linux-gnu and x86_64-linux-gnu.
---
elf/Makefile | 10 +++++++++-
elf/dl-load.c | 8 +++++++-
2 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/elf/Makefile b/elf/Makefile
index 4c9e63dac9..ebb331df36 100644
--- a/elf/Makefile
+++ b/elf/Makefile
@@ -245,7 +245,8 @@ tests += $(tests-execstack-$(have-z-execstack))
ifeq ($(run-built-tests),yes)
tests-special += $(objpfx)tst-leaks1-mem.out \
$(objpfx)tst-leaks1-static-mem.out $(objpfx)noload-mem.out \
- $(objpfx)tst-ldconfig-X.out
+ $(objpfx)tst-ldconfig-X.out $(objpfx)tst-rtld-help.out \
+ $(objpfx)tst-rtld-help-2.out
endif
tlsmod17a-suffixes = 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
tlsmod18a-suffixes = 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
@@ -1907,3 +1908,10 @@ $(objpfx)list-tunables.out: tst-rtld-list-tunables.sh $(objpfx)ld.so
$(evaluate-test)
tst-dst-static-ENV = LD_LIBRARY_PATH='$$ORIGIN'
+
+$(objpfx)tst-rtld-help.out: $(objpfx)ld.so
+ $(test-wrapper) $(rtld-prefix) --help > $@; \
+ $(evaluate-test)
+$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out
+ grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \
+ $(evaluate-test)
diff --git a/elf/dl-load.c b/elf/dl-load.c
index 376a2e64d6..2f760503c5 100644
--- a/elf/dl-load.c
+++ b/elf/dl-load.c
@@ -758,8 +758,14 @@ _dl_init_paths (const char *llp, const char *source,
max_dirnamelen = SYSTEM_DIRS_MAX_LEN;
*aelem = NULL;
- /* This points to the map of the main object. */
+ /* This points to the map of the main object. If there is no main
+ object (e.g., under --help, use the dynamic loader itself as a
+ stand-in. */
l = GL(dl_ns)[LM_ID_BASE]._ns_loaded;
+#ifdef SHARED
+ if (l == NULL)
+ l = &GL (dl_rtld_map);
+#endif
assert (l->l_type != lt_loaded);
if (l->l_info[DT_RUNPATH])
next reply other threads:[~2021-03-15 10:03 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-15 10:03 Florian Weimer [this message]
2021-03-15 12:10 ` H.J. Lu
2021-03-15 12:14 ` Florian Weimer
2021-03-15 12:24 ` H.J. Lu
2021-03-15 12:28 ` Florian Weimer
2021-03-15 12:48 ` H.J. Lu
2021-03-15 12:55 ` Florian Weimer
2021-03-15 12:49 ` Andreas Schwab
2021-03-15 12:55 ` Florian Weimer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87ft0wybrj.fsf@oldenburg.str.redhat.com \
--to=fweimer@redhat.com \
--cc=libc-alpha@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).